Male sexual coercion, in which non-physical tactics are utilized to manipulate unwilling females into sexual contact, is a widespread problem that poses a serious threat to women's health. Despite the prevalence of sexual coercion, etiological studies have most often combined sexual coercion with more severe forms of sexual offending involving the use of physical force. One consequence of "lumping together" these groups is that potentially important distinctions between them may be obscured. It may be, for example, that the etiological factors associated with the use of purely verbal tactics differ from those predictive of more aggressive sexual offending (e.g., forcible rape). If this is the case, the identification of such factors may have important implications for interventions aimed at reducing the prevalence of sexual coercion. The overall objectives of the proposed project are to: (1) develop a multifactor etiological model of male sexual coercion, and (2) to contrast that model with a model of sexual aggression, in order to identify risk factors that may be unique to each type of offending. Using logistic regression techniques, a comprehensive set of attitudinal, personality, behavioral, and child abuse factors will be included in model development. In contrast to the broader sexual coercion literature which has utilized mostly college samples, the present study will be conducted with incarcerated males. The increased base rates of coercion and aggression expected in this population should more easily permit a direct comparison between risk factors associated with each type of offending. [unreadable] [unreadable]